Skip to main content

AI Ethics & Safety

Alignment, bias, governance, and responsible AI

18 entities indexed

PaperAI Ethics & Safety

Training Language Models to Follow Instructions with Human Feedback

by OpenAI

Presents InstructGPT, which uses Reinforcement Learning from Human Feedback (RLHF) to align GPT-3 with human intent. By fine-tuning on human demonstrations and training a reward model on human preference comparisons, InstructGPT produces outputs that human evaluators prefer to GPT-3 outputs despite having 100× fewer parameters.

rlhfalignmentinstruction-following
57C+
BenchmarkAI Ethics & Safety

RealToxicityPrompts

by Gehman et al. / Allen Institute for AI

RealToxicityPrompts measures the propensity of language model generations to produce toxic content when conditioned on a diverse set of 100,000 naturally occurring prompts extracted from the web. It uses the Perspective API to score generated text on toxicity dimensions.

toxicitygenerationsafety
48C
PaperAI Ethics & Safety

Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision

by OpenAI

This paper explores weak-to-strong generalization, a method for training a powerful AI model using supervision from a weaker one. It serves as an analogy for aligning superintelligent AI with human values. The research shows that strong models can learn beyond their weak supervisors and introduces techniques like auxiliary confidence loss to improve performance.

ai-safetyalignmentsuperalignment
48C
BenchmarkAI Ethics & Safety

ToxiGen

by Hartvigsen et al. / MIT

ToxiGen is a large-scale, machine-generated dataset for evaluating nuanced hate speech detection. It contains over 274,000 toxic and benign statements about 13 minority groups, designed to challenge models to identify implicit toxicity without relying on obvious slurs or surface-level cues.

toxicity-detectionhate-speechimplicit-bias
47C
PaperAI Ethics & Safety

Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training

by Anthropic

Demonstrates that LLMs can be trained to behave safely during normal operation but exhibit unsafe behaviors when triggered by specific conditions—acting as 'sleeper agents'—and that standard safety training techniques including RLHF, supervised fine-tuning, and adversarial training fail to reliably remove these backdoors, sometimes even hiding them deeper.

safetydeceptionalignment
47C
BenchmarkAI Ethics & Safety

CyberSecEval

by Meta AI

CyberSecEval is a benchmark developed by Meta to assess the cybersecurity risks associated with Large Language Models (LLMs). It evaluates a model's propensity to generate insecure code, assist in exploiting vulnerabilities, and facilitate attacks, helping safety teams quantify the dual-use risk of code-capable models.

cybersecurityai-safetyllm-evaluation
46C
PaperAI Ethics & Safety

Scalable agent alignment via reward modeling: a research direction

by DeepMind

This research paper proposes a method for aligning advanced AI systems by using recursive reward modeling. The approach leverages AI assistants to help human evaluators assess complex AI actions, enabling scalable oversight and positioning this technique alongside debate and amplification as key AI safety strategies.

alignmentscalable-oversightreward-modeling
46C
PaperAI Ethics & Safety

Representation Engineering: A Top-Down Approach to AI Transparency

by Center for AI Safety / UC Berkeley

Representation Engineering (RepE) is a top-down AI transparency technique for interpreting and controlling Large Language Models. It uses linear probes on activation differences from contrastive prompts to identify and manipulate high-level concepts like truthfulness and emotion without needing to retrain or fine-tune the model.

interpretabilitytransparencyrepresentation-engineering
46C
SkillAI Ethics & Safety

Prompt Injection Defense

by AaaS

Detects and mitigates prompt injection attacks where malicious inputs attempt to override system instructions or extract sensitive information. Implements input sanitization, instruction hierarchy enforcement, and output monitoring to protect LLM-powered applications.

securityprompt-injectiondefense
44C
SkillAI Ethics & Safety

PII Detection

by AaaS

Identifies and flags personally identifiable information (PII) in text data, including names, addresses, phone numbers, SSNs, and financial details. Supports configurable sensitivity levels, redaction strategies, and compliance reporting for GDPR, HIPAA, and CCPA requirements.

piiprivacydetection
44C
SkillAI Ethics & Safety

Output Validation

by AaaS

Validates LLM outputs against expected schemas, formats, and quality criteria before delivery to end users. Implements JSON schema validation, hallucination checks, citation verification, and automated retry logic for outputs that fail validation.

validationoutput-qualityschema-validation
44C
BenchmarkAI Ethics & Safety

CrowS-Pairs

by Nangia et al. / NYU

CrowS-Pairs is a benchmark dataset for evaluating social bias in masked language models. It contains 1,508 sentence pairs with stereotypical and anti-stereotypical statements across nine bias types. The benchmark measures a model's preference for stereotypical completions using pseudo-log-likelihood scores.

biasstereotypesmasked-lm
42C
SkillAI Ethics & Safety

Guardrail Implementation

by AaaS

Implements programmable guardrails that constrain LLM behavior within defined boundaries. Covers input validation, output format enforcement, topic restriction, factuality checking, and automated intervention when model responses deviate from acceptable parameters.

guardrailssafetyvalidation
41C
BenchmarkAI Ethics & Safety

WinoBias

by Zhao et al. / USC

WinoBias is a benchmark dataset designed to measure gender bias in coreference resolution systems. It consists of sentence pairs where pronouns refer to individuals in stereotyped or non-stereotyped occupations, allowing for the quantification of a model's reliance on gender stereotypes versus grammatical correctness.

biasgender-biascoreference
41C
SkillAI Ethics & Safety

Jailbreak Detection

by AaaS

Detects and blocks jailbreak attempts that try to bypass LLM safety training through adversarial prompting techniques. Uses pattern recognition, semantic analysis, and classifier-based approaches to identify known and novel jailbreak vectors before they reach the model.

jailbreakdetectionsecurity
37D
ToolAI Ethics & Safety

Lakera Guard

by Lakera

Enterprise API for protecting LLM applications against prompt injections and content threats. Provides real-time scanning of inputs and outputs for prompt attacks, PII leakage, and inappropriate content.

prompt-injectioncontent-moderationapi
36D
ToolAI Ethics & Safety

Prompt Armor

by Prompt Armor

API-based protection layer for defending LLM applications against prompt injection and jailbreak attacks. Provides real-time input analysis and filtering with minimal latency impact on AI workflows.

prompt-injectionprotectionapi
28D
ToolAI Ethics & Safety

Vigil

by deadbits

Open-source prompt injection scanner for detecting and preventing attacks on LLM applications. Provides multiple detection methods including similarity matching, canary tokens, and heuristic analysis.

prompt-injectionscanneropen-source
25D